Gene therapy is a promising means to permanently cure many genetic diseases. Adeno-associated virus (AAV) is regarded as one of the safest vectors for clinical gene therapy as it is non-pathogenic, infects both dividing and non-dividing cells, has low immunogenicity and can provide stable transgene expression in quiescent tissues. Some of the most significant limitations preventing greater use of AAV in clinical trials are the inability to accurately predict gene transfer efficiencies between animal models and humans, and the small packaging capacity that places a limit on the number of therapeutic genes that can be delivered and thus the variety of diseases that can be treated. The major hypotheses addressed herein are that: (1) differences between intracellular AAV trafficking and/or uncoating events between the predominate AAV serotypes are responsible for the discrepancies between current pre-clinical animal modeling and patient outcomes in clinical trials; and (2) AAV capsid structures are capable of greater diversity to package larger genomes. The specific aims to be studied are as follows: Aim 1: Determine the mechanism of differential AAV2 & AAV8 transduction in human and mouse cells. The intracellular trafficking of differentially labeled AAV2 and AAV8 vectors will be tracked in real time using live cell confocal imaging in both human and mouse cells to determine the mechanism of differential transduction by these AAV serotypes in different species. Specifically, two hypotheses will be addressed via an array of basic molecular biology techniques on subfractionated cells: a) AAV2 and AAV8 uncoat at different rates and/or locations in mouse and human cells; and b) AAV2 and AAV8 enter the nucleus via a different mechanism in human and mouse cells. Aim 2: Evolve capsids that increase the AAV genome packaging capacity. Highly diverse AAV capsid libraries generated by DNA shuffling will be innovatively screened in vitro with applied selective pressure for the ability to package expanded genomes unlike any methodology attempted to date. Capsid variants demonstrating clinical potential will be extensively characterized and vectorized for proof- of-concept pre-clinical studies in animal models.